The presence of tumor-directed immune responses and absence of alternative modes of therapy make renal carcinoma an excellent model for immunotherapy in humans. Preliminary data from two institutions suggest that polymerized tumor antigens admixed with an adjuvant (PTA/Ad) can induce tumor regression and possibly increase survival in patients with disseminated renal carcinoma. We propose a randomized prospective trial of PTA/Ad versus standard hormonal therapy in patients with measurable metastic lesions and a randomized double-blind trial of PTA/Ad versus adjuvant plus placebo in patients without clinical disease but with a high risk of recurrence. Frequency of treatments, dose of antigen, and the adjuvant will be changed to achieve evidence of general and tumor directed cellular immune reactivity as measured by serial skin tests and in vitro testing. Results of in vitro testing and skin tests will be correlated with disease status and clinical response by multivariate analysis. Statistical evaluation of clinical response, disease-free intervals, and duration of disease stability should allow determination of the efficacy of PTA/Ad immunotherapy in a human tumor model.